Path Guiding Using Spatio‐Directional Mixture Models
Authors:
- 14 December 2021
- journal article
- research article
- Published by Wiley in Computer Graphics Forum
- Vol. 41 (1) , 172-189
Abstract
We propose a learning‐based method for light‐path construction in path tracing algorithms, which iteratively optimizes and samples from what we refer to as spatio‐directional Gaussian mixture models (SDMMs). In particular, we approximate incident radiance as an online‐trained 5D mixture that is accelerated by a D‐tree. Using the same framework, we approximate BSDFs as pre‐trained D mixtures, where is the number of BSDF parameters. Such an approach addresses two major challenges in path‐guiding models. First, the 5D radiance representation naturally captures correlation between the spatial and directional dimensions. Such correlations are present in, for example parallax and caustics. Second, by using a tangent‐space parameterization of Gaussians, our spatio‐directional mixtures can perform approximate product sampling with arbitrarily oriented BSDFs. Existing models are only able to do this by either foregoing anisotropy of the mixture components or by representing the radiance field in local (normal aligned) coordinates, which both make the radiance field more difficult to learn. An additional benefit of the tangent‐space parameterization is that each individual Gaussian is mapped to the solid sphere with low distortion near its centre of mass. Our method performs especially well on scenes with small, localized luminaires that induce high spatio‐directional correlation in the incident radiance.Keywords
All Related Versions
This publication has 27 references indexed in Scilit:
- Anisotropic spherical GaussiansACM Transactions on Graphics, 2013
- Online Expectation Maximisation2011
- On-Line Expectation–Maximization Algorithm for latent Data ModelsJournal of the Royal Statistical Society Series B: Statistical Methodology, 2009
- Adaptive importance sampling in general mixture classesStatistics and Computing, 2008
- Intrinsic Statistics on Riemannian Manifolds: Basic Tools for Geometric MeasurementsJournal of Mathematical Imaging and Vision, 2006
- Importance Driven Path Tracing using the Photon Map1995
- A 5D Tree to Reduce the Variance of Monte Carlo Ray Tracing1995
- Maximum a posteriori estimation for multivariate Gaussian mixture observations of Markov chainsIEEE Transactions on Speech and Audio Processing, 1994
- An Antipodally Symmetric Distribution on the SphereThe Annals of Statistics, 1974
- A Stochastic Approximation MethodThe Annals of Mathematical Statistics, 1951